paper(gqa): add 6-case per-PE memory + comm summary chart

Browse Source

Analytical plot generator + PNG covering the 4 sharding cases from
slide 17 plus the 2 d_head-TP variants (Cases 4, 5). Three panels:
per-PE HBM budget breakdown (Wq + Wk + Wv + Wo + 4.24 GB KV
headroom), per-PE KV memory at S_kv=1M, and per-PE communication
per output token (decode). Numbers match slide 17 (max KV context
~7.1M for the 64-way sharded cases).

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>

...

This commit is contained in:

Mukesh Garg

2026-06-16 16:10:49 -07:00

parent 6b6e29968a

commit 694e0cc9b9

2 changed files with 499 additions and 0 deletions

Show all changes Ignore whitespace when comparing lines Ignore changes in amount of whitespace Ignore changes in whitespace at EOL

Download Patch File Download Diff File Expand all files Collapse all files

src/kernbench/benches/1H_milestone_output/gqa/long_ctx/gqa_4cases_summary.png

BIN

Binary file not shown.

Side by Side After Width:  |  Height:  |  Size: 210 KiB